I have been going to national parks literally every year since I was born. I was raised mostly in Utah and my family and I would make sure to visit different parks at least once or twice a year. As I got older, I started to see that these parks were getting overcrowded and less attractive. Even though I still hold a nostalgic love and respect for many of these parks, my family and I now seek to enjoy nature in more isolated places.
So why did these parks become so increasingly popular? In my opinion, National parks are a great way for people to get out and and enjoy nature in a ‘casual’ sense. Prime example, Zion’s National Park provides great hiking trails for mostly beginner to novice hikers while providing amenities such as resort-like lodging and restaurants. They even established a shuttle transportation system around the park so people wouldn’t have to “hike-to-the-hike.” While it does still offer some opportunities for the more experienced and adventurous, it is now a huge hotspot for casuals.
Although these National Parks have now gotten more and more crowded, I would still like to visit them in the future. The primary purpose of my project was to analyze the parks’ statistics and then find data to determine what time of year would be ideal for me to visit them.
Here we see that Grand Canyon National Park is in the lead by a huge margin, having over 6 million people in 2 consecutive years. It’s also important to note that during approximately 2012 and 2013 all parks, except Arches and Capitol Reef, seemed to have experienced a huge exponential growth in yearly visitation. The sudden exponential decay of visitation rates in 2020 was due to the Covid-19 Pandemic.
## Analysis of Deviance Table
##
## Model: gaussian, link: identity
##
## Response: YearlyVis
##
## Terms added sequentially (first to last)
##
##
## Df Deviance Resid. Df Resid. Dev
## NULL 3527 7320268835787259
## Hikes 1 3135661621514508 3526 4184607214272750
## Distance_from_Major_City.mi 1 234511582791206 3525 3950095631481544
## Size_acres 1 1960253793877555 3524 1989841837603990
I fitted a linear model (estimated using ML) to predict YearlyVis with Hotels, Hikes, Distance_from_Major_City.mi, CampUnits and Size_acres (formula: YearlyVis ~ Hotels + Hikes + Distance_from_Major_City.mi + CampUnits + Size_acres). The model’s explanatory power is substantial (R2 = 0.74). The model’s intercept, corresponding to Hotels = 0, Hikes = 0, Distance_from_Major_City.mi = 0, CampUnits = 0 and Size_acres = 0, is at 5.89e+05 (95% CI [5.11e+05, 6.67e+05], t(3522) = 14.86, p < .001). Results within this model:
The effect of Hikes is statistically significant and positive (beta = 7853.19, 95% CI [6036.71, 9669.68], t(3522) = 8.47, p < .001; Std. beta = 0.52, 95% CI [0.40, 0.64]) This means that typically, the more hikes there are, the more likely people will attend each year.
The effect of Distance from Major City mi is statistically significant and negative (beta = -92286.18, 95% CI [-99022.70, -85549.66], t(3522) = -26.85, p < .001; Std. beta = -1.18, 95% CI [-1.27, -1.09]) The farther the distance from the city, the less people will attend each year.
The effect of Size acres is statistically significant and positive (beta = 2.55, 95% CI [2.46, 2.63], t(3522) = 58.33, p < .001; Std. beta = 1.32, 95% CI [1.28, 1.37]) The larger the park, the more likely people will attend each year.
Standardized parameters were obtained by fitting the model on a standardized version of the dataset. 95% Confidence Intervals (CIs).
Conclusion: The results of the model seem consistent and realistic. Hotels and Camp units were not included in the model because I couldn’t connect dates to them. Furthermore, it’s interesting to point out that Grand Canyon National Park was never the highest or lowest in any of the categories, but still has the highest yearly visitation rate.
My ideal conditions are:
Result: November
I fitted a linear model (estimated using ML) to predict MonthlyVisitor with Temperature (formula: MonthlyVisitor ~ Temperature) and then the same model for Precipitation. The model’s explanatory power is substantial (R2 = 0.63). The model’s intercept, corresponding to Value = 0, is at -2.05e+05 (95% CI [-2.33e+05, -1.77e+05], t(503) = -14.19, p < .001). Within this model:
Result: April or October
Result: March
Result: April or October
Result: April or October
Result: September
Result: April or November
AllTrails: Trail Guides & Maps for hiking, camping, and running. AllTrails.com. (n.d.). Retrieved December 13, 2021, from https://www.alltrails.com/.
U.S. Department of the Interior. (n.d.). Nps.gov homepage (U.S. National Park Service). National Parks Service. Retrieved December 13, 2021, from https://www.nps.gov/index.htm.
National Centers for Environmental Information (NCEI). (n.d.). Retrieved December 13, 2021, from https://www.ncei.noaa.gov/.
Expedia travel: Vacation Homes, hotels, car rentals, Flights & More. Expedia.com. (n.d.). Retrieved December 13, 2021, from https://www.expedia.com/.
Other information was searched though google.com